Commentary: risk factors for gastrointestinal bleeding in NSAID users
Version of Record online: 3 OCT 2012
© 2012 Blackwell Publishing Ltd
Alimentary Pharmacology & Therapeutics
Volume 36, Issue 9, page 904, November 2012
How to Cite
Hawkey, C. J. (2012), Commentary: risk factors for gastrointestinal bleeding in NSAID users. Alimentary Pharmacology & Therapeutics, 36: 904. doi: 10.1111/apt.12035
- Issue online: 3 OCT 2012
- Version of Record online: 3 OCT 2012
- Manuscript Accepted: 14 AUG 2012
- Manuscript Received: 13 AUG 2012
The world of anti-inflammatory drugs is a strange one. Non-selective non-steroidal anti-inflammatory drugs (NSAIDs) kill over a thousand patients, annually through upper gastrointestinal (GI) bleeding. Selective cyclooxygenase (COX)-2 inhibitors should have replaced NSAIDs on formularies because they lack this toxicity, cause no more cardiovascular harm than NSAIDs, and indeed lack the ability of NSAIDs to abrogate the benefits of aspirin. However, their use is actively discouraged.
Faced with this muddled thinking, Pfizer has sought additional selling points for celecoxib. As reported by Lanas and colleagues in AP&T, selective COX-2 inhibitors are less likely to be associated with anaemia, a worthwhile additional benefit. They characterise the fall in haemoglobin as the ‘most frequent component/event for the composite GI primary endpoint’ known as a CSULGIE (Clinically Significant Upper or Lower GI Event).
This is a contentious statement because there is no evidence that the anaemia is due to bleeding or associated with iron deficiency or a low MCV. Moreover, the strongest risk factors they identify (albeit with non-stringent statistics) - C-reactive protein and body mass index - more plausibly suggest suppression of haemopoeisis by chronic inflammation. The authors would be wise to keep an open and critical mind on mechanism.
Although less anaemia (by whatever mechanism) is appealing, it is the dominant focus on and lazy belief in a differential cardiovascular risk that has foreclosed proper evaluation of selective COX-2 inhibitors. Two observations encapsulate the problem. In a recent meta-analysis, direct comparisons actually showed a higher incidence of myocardial infarction with naproxen than celecoxib, but complex statistical adjustments allowed the authors to draw an opposite conclusion. On the basis of that conclusion, the UK's QIPP scheme now rewards GPs for increasing proportional prescribing of naproxen or ibuprofen.
Although differences in anaemia are interesting, in a situation as unsatisfactory as this, Pfizer's responsibility is to give continued support to the large trials (PRECISION and SCOT) that can finally provide an answer to this question. They should not waver.
Declaration of personal interests: CJH is a principal investigator on the SCOT trial. Declaration of funding interests: None.
- 4Risk factors associated with a decrease ≥2 g/dL in haemoglobin and/or ≥10% haematocrit in osteoarthritis patients taking celecoxib or a nonselective NSAID plus a PPI in a large randomised controlled trial (CONDOR). Aliment Pharmacol Ther 2012; 36: 485–92., , , et al.